Title :
Classification of grassland types in ibet by MODIS time-series images
Author :
Wen, Qingke ; Liu, Shuo ; Zhang, Zengxiang ; Qiao, Wei
Author_Institution :
Inst. of Remote Sensing Applic., Chinese Acad. of Sci., Beijing
fDate :
June 30 2008-July 2 2008
Abstract :
As one of the five biggest pasturing areas of China, natural grassland in Tibet Autonomous Region, accounts for about 21% of the total area of Chinese natural grassland. Classification of rangeland types is a basic and significant study in stockbreeding. Utilizing the advantages of high temporal resolution of MODIS to construct time series EVI during the grass growth period, dividing large study area to individual regions via altitude and latitude, the paper classifies grassland in Tibet Autonomous Region to 6 types, meadow steppe, typical steppe, desert steppe, high-cold meadow steppe, high-cold typical steppe and shrub herbosa, successfully. This work is one part of the project-land cover mapping of China based on remote sensing images We provide land managers with map of the grassland types and area value of each grassland type in Tibet Autonomous Region in 2005. Average EVI of each grassland types during growth period is induced to reflect relative grass biomass among each type. Shrub herbosa has the biggest average EVI, followed with meadow steppe, high-cold meadow steppe, typical steppe, high-cold typical steppe. Desert steppe has the lowest average EVI.
Keywords :
image classification; time series; vegetation mapping; AD 2005; MODIS high temporal resolution data; MODIS time series images; Tibet autonomous region; average EVI; desert steppe; enhanced vegatation index; grass growth period; grassland type classification; high cold meadow steppe; high cold typical steppe; land cover mapping; natural grassland; rangeland type classification; relative grass biomass; remote sensing images; shrub herbosa; stockbreeding; time series EVI; Biomass; Geography; Geoscience; MODIS; Project management; Remote monitoring; Remote sensing; Soil; Testing; Vegetation mapping;
Conference_Titel :
Earth Observation and Remote Sensing Applications, 2008. EORSA 2008. International Workshop on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-2393-4
Electronic_ISBN :
978-1-4244-2394-1
DOI :
10.1109/EORSA.2008.4620333